Polymer Nanocomposites - ACS Publications - American Chemical

Chapter 5

Surface-Iniatiated Anionic Polymerization: Tethered Polymer Brushes on Silicate Flat Surfaces 1

1

2

1

1

Qingye Zhou , Y o Nakamura , , Seiji Inaoka , Mi-kyoung P a r k , Yingfan Wang , Xiaowu Fan1, Jimmy Mays , and Rigoberto Advincula , 1

1

Downloaded by COLUMBIA UNIV on September 1, 2012 | http://pubs.acs.org Publication Date: November 6, 2001 | doi: 10.1021/bk-2002-0804.ch005

1

*

1

Department of Chemistry, University of Alabama at Birmingham, Birmingham, A L 35294-1240 2Current address: Department of Macromolecular Science, Osaka University, Toyonaka, Osaka 460-0043, Japan

Recently, there has been much interest in using Surface Initiated Polymerization (SIP) in preparing organic-inorganic delaminated nanocomposites. These nanocomoposites have generated much interest because of their potential for applications in high barrier coatings, electronic materials, catalysis, and fundamental studies of polymers in confined environments. Polymer clay/silicate composites have been investigated for a number of years. One of the best ways to achieve control of their application properties is by direct covalent attachment of polymer chains to the particle surfaces using appropriate polymerization methods. Here we report the formation of polystyrene films grafted from silicate surfaces by the living anionic polymerization of styrene. Activated 1,1-diphenylethylene (DPE)-chlorosilane initiation sites were formed by self-assembled monolayers (SAM) on flat silicate surfaces. We have characterized these films by contact angle measurements, atomic force microscopy, attenuated total

© 2002 American Chemical Society

In Polymer Nanocomposites; Krishnamoorti, R., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2001.

39

40 reflection FT-IR, X-ray photoelectron spectroscopy, and ellipsometry. Our results indicate the formation of ultrathin films with unique morphologies and brush densities, as compared to previously polymerized films made by free-radical mechanisms. Thus, they provide insight to differences in anionic versus radical polymerization in surfaces and facilitate new possibilities for block and graft copolymerization from surfaces in the future.


Introduction Surface Initiated Polymerization (SIP) is a method of preparing an assembly of tethered polymer chains in mutual proximity forming so called "polymer brushes". A huge amount of theoretical work has also been devoted on tethered polymer brushes in comparison to the limited experimental information available on their properties and structure. (1,2,3) This unique geometry at solid interfaces ideally results in end-grafted, strictly linear chains of the same length where the grafting density is sufficiently high with respect to the equilibrium radius of gyration (R ) of the grafted macromolecules. (4) To avoid steric crowding, the polymer chains are forced to stretch away from the interface, resulting in a brush height (h), which is significantly larger than R . Tethered polymers on flat surfaces have involved polymer physisorption and chemisorption (covalent attachment). For polymer physisorption, diblock copolymers are used where one block strongly adsorbs to the surface. The usually shorter "anchor" block adsorbs strongly onto the surfaces, leaving the remaining "buoy" block, tethered to the interface.(5) For covalent attachment or chemisorption, tethering has been accomplished by grafting preformed polymers to tethering sites, a "grafting to" approach or by polymerizing from surfaceimmobilized initiators, a "grafting from" approach. These differences are shown schematically in Figure 1. Examples of the latter approach include polymerization using surface-immobilized azo-bis isobutyronitrile (AIBN) analogues,(6) atom-transfer radical polymerization (ATRP),(7) 2,2,6,6tetramethylpiperidinooxy (TEMPO),(£) cationic (9) and anionic(iO) initiators. The "grafting from" methods have intrinsic advantages in that the grafting density is not self-limiting, i.e. the initiator is bound to the surface and is primarily dependent on monomer diffusion and reactivity. With the "grafting to" approach, as the grafting density increases, the chains have to stretch to allow further grafting, which results in a decrease in the rate of grafting kinetics. (11) g

g


41


Grafting to

Grafting From

Figure 1. Schematic diagram comparing the grafting to and grafting from approaches with chemisorbed macro molecules. Spheres represent reactive endgroup or initiator. Organic-inorganic nanocomposites have generated much interest because of their potential for applications in tough and high temperature-compatible, particle-reinforced polymers, coatings, electronics, catalysis, and the study of polymers in confined environments. (12) One important class of organic/inorganic hybrids that has been developed over the past few years are polymer-layered silicate (PLS) nanocomposites. (13,14,15) These nanocomposites have generated much interest because of the need for new polymeric materials and composites with suitable barrier properties, e.g. for soldier protection. This includes chemical resistance, selective permeability, resistance to environmental changes, and reduced weight with increased strength. Layered silicates (mica, clay) have been "blended" with polymers to produce two types of structures: 1) intercalated structures where a single extended polymer chain is positioned between the silicate layers, resulting in alternating layers of the polymer and the inorganic sheets, with a repeat distance of a few nanometers; 2) delaminated structures where the silicate layers are exfoliated and dispersed in a continuous polymer matrix, more like a traditional filled polymer system. A group at the Toyota Research Center in Japan first demonstrated the remarkable properties of PLS nanocomposites using the nylon/montmorillonite system.(13) They showed that addition of low levels of clay (1-10% by weight) doubled the modulus and tensile strength and increased the heat distortion temperature by 100 °C. These property improvements allow the use of these composites in higher temperature applications (e.g. under the hood in cars). Also, because only a very small amount of the clay is necessary to achieve these property enhancements, the composites are very lightweight. Furthermore, the plate-like nature of the clays give two-dimensional reinforcement compared to the one dimensional reinforcement obtained with


42 polymer/fiber composites, and, in the case of the intercalated structures, materials with outstanding barrier properties can be generated. A key factor in all work involving mixtures of polymers with inorganic materials is the nature of the interface. Most of the work on modifying particle surfaces with polymers has focused on physical adsorption of homopolymers or copolymers. However, such adsorbed layers are susceptible to removal, e.g. by exposure to a thermodynamically good solvent. Furthermore, it is often difficult to precisely control the physical structure of such adsorbed layers (thickness, chain orientation, placement of functional groups, etc.). In the case of intercalated nanocomposites incorporating clays, there are difficulties in fabrication due to the close packing of the silicate sheets. This inherent spacing is of the order of lnm, which is much smaller than the R of a typical polymer chain. Thus, there is a large entropic barrier, combined with a strong enthalpic barrier, that inhibits the penetration of the polymer into this region and, consequently the intercalated structure cannot be readily formed by simple adsorption of homopolymers or block copolymers (in melt blending or extrusion processes). The Toyota group overcame this problem and produced composites of clay intercalated with nylon by swelling the clay with caprolactam, which reacts with the charged Montmorillonite surface, followed by addition of more caprolactam or related nylon monomers.(16) While effective with nylon, this is not a method that is widely applicable to various monomers. The polymer is bound to clay by ionic bonds (the common natural type of Montmorillonite is composed of sodium ions and negatively charged Montmorillonite). Vaia and Giannelis have worked extensively with composites of organically-modified layered silicates (OLS).(i4) OLS are composed of surfactant modified clays commonly made via a cation exchange reaction. They have emphasized that whether a mixture of polymer and OLS forms an exfoliated, intercalated, or conventional nanocomposite depends critically on the character of the polymer and the OLS. Enthalpic and entropic considerations between polymer, silicate, and surfactant, needs to be defined. Existing guidelines as to the optimum OLS/polymer combinations are unsatisfactory, making hybrid synthesis a laborious trial and error process. Thus a general method suitable for use with a wide range of monomers/polymers is lacking. Current synthetic methods are incapable of controlling the polymer conformation, graft density, and architecture to produce highly dispersed and processible materials. For delaminated nanocomoposites, the best way to achieve property control is by direct covalent attachment of polymer chains to the particles using an appropriate polymerization method, ( i 7) To address the investigation of polymerizable systems for covalently tethered polymers to clay and silicate surfaces, we have investigated surface initiated anionic polymerization. In this article, we report the formation of polystyrene films grafted from a silicate (SiOx layer) surface by living anionic


16

g



43 polymerization of styrene. Activated 1,1-diphenylethylene (DPE)-chlorosilane initiation sites were formed by self-assembled monolayers (SAM) on silicate surfaces (SiOx layer of a silicon wafer). To our knowledge, the formation of these polymer brushes by the activation of DPE on surfaces has not been previously reported. Previous work on anionic polymerization has focused on other initiators and surfaces .(10,18) We have characterized these films by contact angle measurements, atomic force microscopy (AFM), FourierTransform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), and ellipsometry. As model studies, we need to distinguish the concept of surface initiation from our eventual goal, which is intergallery initiation in silicate (clay) particles. First, it is important to investigate in detail, the anionic polymerization mechanism from ideal (flat silica) surfaces.

Experimental Synthesis of the Initiator The initiator was synthesized: 1,1-diphenylethylene (DPE)-chlorosilane as shown in the schematic diagram in Figure 2. The DPE is separated from the silyl group by an alkyl spacer (8 and 11 carbon lenghts were synthesized). The intermediate products were analyzed primarily by N M R were found to be consistent with the reported structures and constitution. DPE was specifically chosen to avoid self-polymerization parallel to the plane surface as attached to the surface by SAM.(IP) It is then reacted in-situ with sec-BuLi or n-BuLi to activate it for anionic polymerization. Synthesis of 4-Bromo-DPE (I). To a 500 mL round-bottomed flask, fitted with a dry nitrogen inlet septum, methyltriphenylphosphonium iodide (31 g, 76 mmol) was suspended in dry THF under nitrogen gas atmosphere. To the suspension was added n-BuLi (31 mL of 2.5 M in hexane, 76 mmol) at room temperature with stirring. The mixture immediately became dark red and was stirred continuously for half one hour. 4bromobenzophone (20 g, 76 mmol) was then added to the mixture by syringe over a 30 minute period with vigorous stirring at room temperature. After completion of the addition step, the mixture became yellow and was stirred overnight at room temperature under nitrogen atmosphere. It was then diluted with 150 mL of chloroform and 150 mL of diluted hydrochloride acid aqueous solution. The organic phase was collected, washed and dried over MgS(>4. The solvent was removed by rotary evaporation and the resulting residue purified by chromatography on silica gel using n-hexane as the eluent. This procedure


44

resulted in 16.8 g of 4-bromo-DPE as yellow oil (yield). The product can also be purified by distillation. H N M R (CDC1 ) 8 7.44 (2H, d, J = 8.7, Ar-H), 7.30 (5H, m, Ar-H), 7.19 (2H, d, J = 8.7, Ar-H), 5.44 (2H, d, / = 3.0, C=CH ). C NMR(CDCl ,300MHz): 149.5, 141.4, 140.9, 131.8,130.4, 128.8, 128.7, 128.4, 122.3,115.2. l

3

l 3

2


3

Figure 2. Schematic diagram of initiator synthesis (DPE) functionalized with silane coupling agent and alkyl spacer.

Synthesis of4-(ll '-undeceneyl)-DPE (2) 1.5 g (51.8 mmol) of activated magnesium turnings and a small iodine particle were placed in a three-necked flask equipped with a reflux condenser and purged with nitrogen. A portion of a solution of 11 -bromo-1 -undecene (12 g, 51.8 mmol) in 20 mL of dry diethyl ether was then added to the flask. Following the initiation of the Grignard reaction, the rest of the 11-bromo-1undecene solution was added over 30 min. After refluxing for 2h, the Grignard reagent was transferred to an addition funnel, and was slowly added to a mixture of 40 mg of 1,3-bisdiphenylphosphinopropane Nickel (II) chloride or Ni(dppp)Cl and 5 g (19.3 mmol) of 4-bromo-DPE in 20 mL of diethyl ether. After stirring at room temperature for 12 hrs, the mixture was acidified with dilute hydrochloride aqueous solution and extracted with diethyl ether three times. The combined organic extracts were washed with a saturated aqueous Na2C0 solution and dried over anhydrous Na S04. Distillation under reduced pressure gave 3.92 g of 4-(undec-9-enyl)-DPE as a colorless oil. H N M R (CDCI3) 8 7.34-7.32 (5H, m, Ar-H), 7.24 (2H, d, J = 8.7, Ar-H), 7.13 (2H, d, J= 2

3

2

l


45 8.5, Ar-H), 5.44-5.41 (1H, m, -CH=), 5.40 (2H, d, J = 3.0, =CH ), 5.03-5.01 (2H, m, =CH ). 2.61 (2H, t, J = 6.6, - C H ) , 2.05-1.28 (m, 16H, -(CH ) -). C N M R (CDCl ,300MHz) 150.4, 143.0, 142.2, 139.7, 139.1, 128.8, 128.6, 128.5, 128.0, 114.6,114.1, 36.1, 34.3, 33.1, 31.9, 29.9,29.8,29.7, 29.6, 29.4. 2

l 3

2

r

2

8

3

Synthesis of chloride silane-DPE derivative (3). 4-(l l'-undeceneyl)-DPE (2) (1.73 g, 5.94 mmol) was dissolved in 10 mL of dry toluene followed by addition of (851.5 mg, 9mmol) of chlorodimethylsilane and three drops of the catalyst Platinum(0)-l,3-divinyl-l,l,3,3tetramethyldisiloxane complex in xylene with N protection. After the mixture was heated overnight under N atmosphere at 50-60 °C, the solvent and unreacted chlorodimethylsilane were removed under vacuum. H N M R analysis indicated that all 4-(ll'-undeceneyl)-DPE (2) was completely consumed. Yield 1.82g. H N M R (CDC1 ) 8 7.34-7.32 (5H, m, Ar-H), 7.24 (2H, d, J = 8.7, ArH), 7.13 (2H, d, J= 8.5, Ar-H), 5.40 (2H, d, / = 3.0, =CH ), 2.61 (2H, t, J = 6.6, -CH -), 2.05-0.71 (m, 16H, -(CH ) -), 0.29 (s, 6H, -Si(CH ) -).


2

2

l

l

3

2

2

2

10

3

2

Self-assembled Monolayers of the Initiator Self-assembled monolayers of the DPE initiator were prepared as follows: Both glass and silicon wafer substrates were plasma (under Ar, using PLASMOD, March Instruments) and piranha solution (30% H 0 : 70%H SO , 15 min) cleaned. The substrates were thoroughly rinsed with Milli-Q purified H 0 and dried in an oven (90 °C) overnight before use. Modified gold-coated quartz crystals (QCM substrates) were plasma treated under 0 conditions.XX To adsorb the DPE initiators, the cleaned substrates were immersed in 1 m M of the organosilane initiator in toluene solution (under N gas) for more than 12 hours. After reaction, the substrates were sonicated for 15 min. with 1) toluene, 2) toluene: acetone, 1:1 and 3) acetone in sequence. The substrates were then immediately transferred to the vacuum line (under N gas) and sealed. Similar procedures for cleaning and adsorption were adapted for the self-assembly of the initiator on modified gold surfaces. The surfaces were characterized before and after polymerization. 2

2

2

4

2

2

2

2

Characterization of the Surface and Polymer Films One major objective is to compare anionic polymerization on surfaces and their unique properties compared to solution polymerization. The surface properties and chemistry involved in the initiation process may dramatically affect a living polymerization mechanism. Characterization of the surface is



46 critical in determining the kinetics, mechanism, and stability of surface initiated polymerization. We have used: Atomic force microscopy (AFM)-Molecular Imaging (PicoScan) to investigate surface morphology. This involved the use of Magnetic-AC or M A C modes to minimize tip-surface interaction and sample deformation. The M A C lever consists of a Silicon Nitride based cantilever coated with magnetic films. The force constant is about 0.5 N/m with resonant frequency in air between 90 to 105 kHz. The sample stage has a solenoid mounted underneath the plate. Ellipsometry for thickness measurements was done using a Microphotonics SE 400 Sentech Ellipsometer at and incidence angle of 70° (He-Ne laser, 632.8 nm) using a refractive index value of 1.45 for the S A M and 1.50 for the polymer. Contact angle measurements, both dynamic and static measurements, were done on a MicroCAM contact angle goniometer and gave data for surface energy and wetting characterizations. The measurements were taken at several different spots of the same substrate and the average value reported. Quartz Crystal Microbalance measurements were done on a M A X T E K , Q C M system with a PM-740 frequency counter. The Sauerbrey equation was used to relate the mass increase (due to chemisorption) with the frequency change.(20) Preliminary cyclic voltammogram measurements were done on the quartz substrate with a Bioanalytical Systems BAS-100B Electrochemical Analyzer. This was done with a Ag /Ag+ working electrode and platinum wire in anhydrous acetonitrile (Aldrich) containing 0.1 M LiClC>4 as supporting electrolyte. Cyclic voltammograms were determined at 100 mVs" and were integrated graphically. XPS measruements were made with a Kratos Axis 165 SAM/XPS system (University of Alabama at Tuscaloosa). Size exclusion chromatography (SEC) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) were used to analyze the polymers formed in solution. FT-IR ATR measurements were made with the Bomem Prota FT-IR using an variable ATR attaehement (Spectratech Model 300) with KRS-5 prism (2.37 R.I.).

1

Polymerization Procedure for Polystyrene Styrene was stirred with CaH overnight, degassed 3 times on the vacuum line, and then distilled into the flask containing dibutylmagnesium . After stirring for a few hours, it was distilled into ampoules. Benzene was stirred with CaH overnight, degassed 3 times and then distilled into a flask with BuLi and a small amount of styrene (orange color). The high vaccum apparatus is shown in Figure 3. After placing the SAM-coated glass or silicon wafer (with DPE) in the reactor and sealing, the s-BuLi or n-BuLi (about 1 x 10" mol in 5 mL hexane) was injected into the purge section through the septum. The solution was left with the substrate overnight. Excess BuLi solution was then removed and the surface was washed several times with degassed benzene to ensure removal of excess BuLi 2

2

3


47 and impurities. THF or BuOLi was also added into the initiator solution to observe its effect on the initiation of DPE. The styrene monomer was then introduced by breaking the break seal between the reactor and the monomer ampoule. The polymerization was allowed to proceed for periods of a few hours to a few days before it was terminated by MeOH. The samples were taken out of the reactor and immediately washed using Sohxlet extraction procedures for 36 hours. A complete description of the polymerization and characterization results will be reported in a forthcoming publication. (21)


vacuum e line

iw^eak seal

reactor silicon waiter purge; section

Figure 3. Schematic Diagram of the polymerization set-up under high vacuum and inert conditions. A: ampoule containing THF, B: ampoule containing methanol, C: ampoule containing styrene.

Results and Discussion Synthesis and Immobilization of the Initiator The design parameters of the initiator combine: DPE functionality, spacer (alkyl chain), silyl agent (chlorosilane group), and capacity for postpolymerization cleavage. The DPE is separated from the silyl group by an alkyl spacer. DPE was specifically chosen to avoid self-polymerization parallel to the plane surface. The DPE-chlorosilane was then attached to the surface by S A M . The tethered DPE moieties were then reacted with BuLi for activation. The S A M preparation work was done under vacuum or under inert atmosphere (dry box) as appropriate. For controlled dispersion of the grafting sites, variation of initiator density; mixtures and co-adsorption of the DPE derivative with varying amounts of alkyldimethylchlorosilane was used to control the DPE density on the surface

American Chemical Society Library 1155 16th S L HW. In Polymer Nanocomposites; Krishnamoorti, Washington, 20Q3B R., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2001.

48

while eliminating the Si-OH groups that can interfere with the polymerization.(22) Ellipsometric measurements on the Si wafer indicated a thickness of about 1.7 nm. The contact angle measurements showed a change from 20° to 82°, reflecting the change in the wetting properties of the surface after chemisorption of the initiator by S A M . A F M measurements did not indicate any clear morphology features compared to that of bare SiOx surface. We also characterized the adsorption properties of the initiator on modified gold electrode and quartz crystal surfaces.(25) Using Q C M , a change of AF=-144 (Hz) indicated a thickness of 1.7 nm for the DPE initiator assuming a density of 1 g/em . This is consistent with that obtained by ellipsometry. Using cyclic voltammetry experiments, the DPE pendant groups on the monolayer can be oxidized on a conducting gold electrode surface to quantitatively asses the molecular density of the monolayer. A schematic diagram of the oxidation process is shown in Figure 4. This assumes that 1 electron per molecule is consumed for coupling and 0.5 electron per molecule will participate in the redox reaction (a total of 1.5 eeleetron/moleeule will be consumed at the very first cycle. Preliminary cyclic voltammogram measurements indicated that DPE functional groups showed oxidation with the onset of +1000 mV(vs A g /Ag+). The oxidation potential is lower than the background oxidation of the solvent (acetonitrile). Based on the area of the electrode (50 mm ), electron density (16 A/electron), and the charge injected (40-80 (iC), it was reasonable to conclude that a monolayer of DPE can be formed on the surface.(2¥) A molecular density value of 24.1 A/molecule was obtained which is typical of a well-packed S A M monolayer bearing electroactive species.


3

2

Figure 4. Schematic Diagram of the oxidation onset on the SAM of the DPEinitiator on a QCM crystal (gold electrode) indicating monolayer formation.


Model surfaces of float glass and mica are also currently being investigated. These substrates are idealized surfaces appropriate for several surface sensitive spectroscopic and microscopic techniques. It is important to compare their properties and differences with clay and silica particulate surfaces.(25)


Polymerization and Characterization at the Surface Using the polymerization procedure, several attempts were made with samples prepared inside the constructed vacuum reaction vessel (SIP B-3, B-4, B-5, S-3 and S-4) and in an inert atmosphere glovebox (SIP-B-9 and SIP-B-6). The polymerization was activated by introducing n-Bu-Li or S-BuLi, followed by the addition of the monomer. This is shown schematically in Figure 5. After polymerization, all the films were washed for more than 36 hours by Soxhlet extraction in toluene prior to characterization of the surfaces. The following polymerization conditions and results were obtained and summarized in Table 1. CH-

—OH Cl-SiMc -(CH ) ' 2

2

u

—0-SiMc (CH ), r

2

Bu.

0^ —O— SiMc -(CH )ir 2

2

toluene Bu.

,0

©

Li

—O— SiMc -(CH )i 2

2

growing polymer chain

Figure 5. Immobilization of the DPE initiator and activation by the addition ofsec-Buli. Monomer is eventually added and results in polymerization.


50

The data in Table 1 indicates several attempts at polymerization on silicon wafers (attached to glass). The best repeatable attempts obtained films of 16.1 ± 0.2 nm average thickness for dry polystyrene brushes as obtained using ellipsometry. These results are comparable to that obtained by Ulman and coworkers (18 ± 0.2 nm), where they reported homogenoues films of uniform eoverage.(i#) They have estimated a degree of polymerization of N=382 and a grafting density of 7-8 chains/i?/ or 3.2-3.6 nm /chain based on mean-field theory calculations from in-situ swelling experiments, monitored by ellipsometry. The use of n-Buli resulted in better polymerization, as it was easier to remove by washing compared to s-BuLi. This resulted to lower amounts of free initiator species (BuLi) present that may initiate the polymerization of styrene from solution. We have observed this in cases where unremoved BuLi is present in solution. This resulted in polymers with up to Mn = 100 x 10 , P.D =1.44, as analyzed by SEC and MALDI. THF was added together with nBuLi to accelerate the initiation of DPE. We also did not observe any significant improvement with samples prepared (SAM and polymerization) using a glove box procedure with inert atmosphere conditions. Attenuated total reflection (ATR) FT-IR measurements verified the presence of grafted polystyrene based on the functional groups present, 3,100-2,700 cm" aliphatic C H and 1506, 733 cm* C=C aromatic stretching vibrations (upon comparison with standard polystyrene peaks). Polarized FT-IR measurements will be conducted to determine the preferential orientation of the polystyrene chains.


2

4

1

1

Table 1. Polymerization Conditions and Results SAMPLE SIP-B-3 SIP-B-4 3 days SIP-B-5 5 days SIP-S-3 (3 days) SIP-S-4 (5 days) SIP-B-9 SIP-B-6

Activation Additive s-BuLi n-BuLi

none THF

n-BuLi

THF

s-BuLi

BuOli

n-Buli

THF

n-BuLi n-BuLi

THF THF

Thickness in nm * 6.8 10.8 (16%) 16.1 (100%) 13.4(16%) 13.7(100%) 3.8(16%) 3.9(100%) 6.3 (16%) 8.2 (100%) 8.9 4.7

Contact Angle 95 70 76 95 89 50 72 61 64 82 74

NOTE: * brackets indicate % of DPE initiator with alkydimethylchlorosilane solution (0.0001M) for SAM.



51 We have investigated the wetting properties of these films. The water eontact angle values using the sessile drop is summarized in Table 1. We have found the values to be lower than previously reported values for bulk polystyrene but comparable to previously reported grafted polystyrene brushes of similar thickness. These were verified with hysteresis measurements (advancing and receding). However, no correlation can be made with regards to polymerization time and solvent on the wetting properties and thickness of the films. Likewise the use of 16% vs 100% DPE did not show any clear differences in either contact angle or thickness measurements. This is perhaps due to the poor monolayer mixture properties (further investigations are being made). Thus, contact angle measurements are inconclusive at this point with regards to polymerization trends and conditions and their effect on brush density and polymer conformation. A F M Investigations To investigate the surface topography of the grafted films, A F M measurements was made. From the different polymerization conditions we have investigated, we found that the dry films (measured in ambient air) have a consistent morphology made up of holes (surrounded by reliefs) statistically distributed within domains, as shown in Figure 6. The smooth relief regions of the film 0.2-0.4 jum, have an rms of 0.5-0.8 nm. The films prepared with the 16% initiator composition showed regions without polymer films. This general morphology is uncharacteristic of previously reported grafted polystytrene systems using other types of intitiators, in particular with the work by Prucker and Ruhe using free radical SIP where they obtained smooth homogeneous films.(26) The group of Ulman and co-workers also reported smooth, homogeneous polymer surfaces throughout the entire substrate on the macroscopic as well as microscopic scale with roughness of 0.3-0.5 nm (rms). (18) However, they have reported smaller "dimples" typically 2-3 nm deep, interesected by small areas of defects. These defects consists of dense arrangements of holes (18-20 nm deep, corresponding to the layer thickness measured by ellipsometry) surrounded by rims. These observations are more similar to the morphologies we have found with these films. Thus, compared to their results, we would expect lower degrees of polymerization and grafting density for these films. However, it is not clear whether the morphology is a result of the polymerization mechanism or is indicative of incomplete initiation or perhaps domain formation after post-polymerization treatment. Since this morphology was consistent for all the samples we investigated, we intend to investigate in detail the properties of the surface initiator and its property before and after


52


polymerization. One hypothesis is the involvement of "physisorbed initiators" being removed after Soxhlet extraction, thereby producing the holes. Another is the property of the "anionic surface" or nucleophilicity and the surface energy involved, affecting the mechanism of chain growth. Clearly, the use of mixed monolayers for variable initiator density needs further investigation. Efforts are underway to study in detail these hypotheses and parameters using XPS, SEM and Q C M and correlate with polymerization results. Thus, we are currently investigating the unusual morphology and correlating it with the characteristics of the surface, density of initiator sites, and surface energy.

Figure 6. AFM Image ofgrafted polymer with a relative height up to 27.5 nm (white). XPS Measurements XPS of the initiator surface before and after polymerization was made. It is invaluable for determining and monitoring changes on the surface functional groups as a result of decomposition or conversion to reactive species for initiation. The spectra are shown in Figure 7. Our results showed the presence of the relevant C peak, representing the presence of the polystyrene film on the surface (SIP-B-5 sample). This peak was absent from the XPS spectra of the Si wafer and was small (by integration) on the spectra of the S A M coated substrate (Figure 7a). It is interesting to note that the presence of relevant Si peaks (99.8 for Si and 103.00 for SiOx) even with the polymerized film indicated the exposure of bare SiOx surfaces. This is consistent with the morphology observed by A F M were we supposed the depressions (holes) observed to be coincident with the silicate surface. The absence of L i peaks (54.9) is indicative that no active anion is present and all the L i salts have been removed by the Sohxlet extraction and washing process for these sample.


53

6000-

Rate

500040003000-


2 O a

20001000— i

1200

1000

1

1—

800

—i

600

1

1

400

1

1

200

1

1

0

Binding Energy [eV] (a)

1200

1000

800

600

400

200

Binding Energy [eV] (b) Figure 7. XPS spectra of the Si wafer with (a) SAM of initiator and (b) the polymer film grafted on the Si wafer.



54

Other parameters of a typical anionic polymerization needs to be investigated: concentration and ratio of monomer, temperature, and solvent. It is important to determine the nucleophilicity of this anionic initiator as compared to other systems. A living polymerization mechanism will allow us to prepare and investigate different block and graft copolymers, end-functionalized polymers, and grafting (branched) architectures. However, is it not possible to obtain more polymer samples using our present methodology. Thus, we are currently conducting polymerization on glass and silica particles (high surface/volume ratio) to obtain high surface areas of grafted polymer. A suitable technique for detachment of the polymer chain from these systems will allow us to characterize the molecular properties, including molecular weight, polydispersity, structure, terminal groups, structure analysis, tacticity, block and graft copolymer architecture. Such analysis can only be completely analyzed when the polymer is detached from the surface. Our fundamental research goal at this point is in understanding of the living surface initiated polymerization; this necessitated the initial use of well-characterized and idealized surfaces. Once we have achieved this objective, we will be applying the methodology to other actual inorganic (clay) surfaces and particulates.

Conclusion In this article, we have presented our work on the surface initiated anionic polymerization of styrene using a S A M of DPE modified silane initiators on Silicate surfaces. The results indicate the formation of polymer brushes with a unique morphology not typical with analogously observed polymer brush systems. We have used a combined approach of spectroscopic and microscopic surface analysis to probe the system. In the future, we would like to understand the mechanism and efficiency of this initiator system compared to others and to use this initiator on actual silica and clay particles. More importantly, we intend to demonstrate the feasibility of forming block mid graft copolymers by surface initiated living anionic polymerization. Acknowledgment Funding for this project from the Army Research Office (ARO) under DAAD19-99-1-0106 is gratefully acknowledged. We acknowledge the XPS measurements by Prof. Greg Szulczewski of the Department of Chemistry, University of Alabama at Tuscaloosa and the use of the electrochemistry set-up with Dr. Juan Pablo Claude (UAB). We also acknowledge Dr. Shuangxi Wang for the re-synthesis of the DPE initiator.


55


References 1. Wittmer, J. P.; Cates, M . E.; Johner, A. and Turner, M. S. Europhys. Lett. 1996, 33, 397. 2. de Gennes, P.-G. J. Phys. (Paris) 1976, 37, 1443. 3. de Gennes, P.-G. Macromolecules 1980, 13, 1069. 4. Sanchez, I. C. Physics of Polymer Surfaces and Interfaces; Butterworth: London, 1992. 5. Dan, N . ; Tirrell, M . Macromolecules 1993, 26, 4310. 6. Prucker, O.; Ruhe, J. Macromolecules 1998, 31, 602-613. 7. Ejaz, M . ; Yamamoto, S.; Ohno, K.; Tsujii, Y.; Fukuda, T. Macromolecules 1998, 31, 5934-5936. 8. Husseman, M . ; Malmstrolm, E.; McNamara, M . ; Mate, M . ; Mecerreyes, D.: Benoit, D.; Hedriek, J.; Mansky, P.; Huang, E.; Russell, T.; Hawker, C., Macromolecules 1999, 32, 1424-1431. 9. Jordan, R. and Ulman, A. J. Am. Chem. Soc 1998., 120, 243. 10. Oosterling, M . L. ; Sein, A.; Schouten, A. J. Polymer 1992, 33, 4394. 11. Dorgan, J. R.; Stamm, M.; Toprakcioglu, C.; Jerome, R.; Fetters, L. J. Macromolecules 1993, 26, 5321. 12. Giannelis, E. P. Adv. Mater. 1996, 8, 29. 13. Usuki, A . et al. J. Mater. Res., 1993, 8, 1174 ; ibid, 1993, 8, 1179; J. Polym. Sci., Polym. Chem., 1993, 31, 2493; ibid,1993, 31, 983; J. Polym. Sci., Polym. Phys. Ed., 1994, 32, 625 ; ibid, 1995, 33, 1039. 14. Vaia, R. A. ; Giannelis, E. P. et al., in Synthesis and Processing of Ceramics: Scientific Issues, W. E. Rhine, M. T. Shaw, R. J. Gottshall, and Chen, Y. (Eds.), MRS Proceedings, Pittsburgh, PA, 1992; Chem. Mater. 1993, 5, 1694; ibid, 1994, 6, 1017; J. Am. Chem. Soc.1995, 117, 7568; Macromolecules,1995 28, 8080 ; Chem. Mater.1996, 8, 1728; Macromolecules 1997, 30, 7990; ibid, 1997, 30, 8000. 15. Ginzburg, V . and Balazs, A. Macromolecules 1999, 32, 5681 16. Fleer, G. J. ; Cohen-Stuart, M . A . ; Scheutjens, J. M . H . M. ; Cosgrove, T. and Vincent, B. Polymers at Interfaces, London: Chapman and Hall, 1993. 17. Weimer, M . ; Chen, H.; Giannelis, E.; Sogah, D. J. Am. Chem. Soc. 1999, 121, 1615-1616 18. Jordan, R.; Ulman, A ; Kang, J.; Rafailovich, M . ; Sokolov, J. J. Am. Chem. Soc. 1999, 121, 1016-1022 19. Rempar, P. and E. Merrill, Polymer Synthesis, 2nd Ed. Uthig and Wepf, Heidelberg, 1991. 20. Sauerbrey, G. Z. Physik, 1959, 155, 206. 21. Zhou, Q.; Nakamura, Y.; Inaoka, S.; Park, M . ; Mays, J.; Advincula, R. in preparation. 22. Wirth, M . ; Fairbank, R.; Fatunmbi, H. Science, 1997, 275, 44. 23. Taylor, D.; Morgan, H,; D'Silva, C. J. Phys. D: Apply. Phys. 1991, 24, 1443. 24. Inaoka ,S. and Collard, D. Langmuir 1999, 15, 3752 25. Van Olphen, H. An Introduction to Clay Colloid Chemistry;Wiley Interscience: New York, 1977; 2nd. Ed. 26. Prucker, O.; Ruhe, J. Langmuir 1998, 14, 6893-6898.


Polymer Nanocomposites - ACS Publications - American Chemical

Recommend Documents